摘要
提出一种新的自适应目标图像分割，它适用于复杂环境下的运动图像跟踪。该算法利用序列连续图像的帧间相关性和差异性检测目标，用直方图滤波法分割目标，根据目标的运动特性在相邻帧中确认目标，并尽可能抑制背景噪声的干扰。与某些常规算法不同的是，新方法充分考虑了目标－背景条件，帧间相关性和差异性，目标检测及提取算法分三步：运动能量检测、直方图滤波及快速定位。
An automatic image segmentation method applied to real time image tracking in complex environment is proposed. In accordance with the inter frame correlatively and the histogram of the image, the segmentation can be executed. We use motion characteristics of moving targets for segmentation. Meanwhile, it suppresses the interruption of background noise as much as possible. It is different to conventional algorithms that the new approach proposed integrates proper object background condition, inter frame correlatively and differences into a three step segmentation process which include motion energy detection, histogram filter and a fast locating procedure. The experiments conducted on visible band image sequences are reported.
出处
《铁道学报》
EI
CSCD
北大核心
1996年第6期33-38,共6页
Journal of the China Railway Society
